EDUCAUSE QUARTERLY Number 3 2005

40
An ERP
Post-Implementation
Review:
E
nterprise resource planning (ERP)
systems have become a fact of
life in higher education. A 2002
EDUCAUSE Center for Applied Research
(ECAR) study estimated that higher edu-
cation had spent close to $5 billion on
ERP systems.
1
According to the study,
54 percent of survey respondents had
implemented an ERP as of the summer
of 2002, and an additional 35 percent
expected to implement an ERP module
or to add a module to an existing system
within the next three years. By the time
this article appears in printassuming
these plans were realizeda large major-
ity of institutions of higher education
will be running ERP systems.
An ERP systems success rests on the
integration of data across the institu-
tion. An ERP eliminates the need for
individual data stores, duplicate records,
and coordination of disparate data sys-
tems with unique record formats. At one
institution, circumstances that argued
for moving to an ERP system included
a redundant, disorganized database
structure;
inaccurate data;
difculty in reporting and sharing
information;
dependence on manual processes and
human interventions;
problems in providing seamless cus-
tomer service between ofces;
difculty complying with reporting
requirements;
heavy reliance on the computing cen-
ter staff; and
lack of capacity for process improve-
ments.
2
Many participants at an EDUCAUSE
2001 Roundtable cited an increased
ability to provide student services as
an additional reason for moving to an
integrated ERP.
3
ERP systems promise to
increase operational efciency, improve
customer service, and help enforce an
institutions business rules.
An ERP system represents a signi-
cant investment, both for acquisition
and for ongoing support. At small
and mid-size institutions, the cost
of acquiring an ERP system can seem
daunting. For many such institutions,
A study of Gonzagas use of its enterprise resource planning
system recommended further investment in the software
By Wayne D. Powel and Jim Barry
LOOKING BACK
Planning for the Future by
Number 3 2005 EDUCAUSE QUARTERLY
41
the ongoing support costs, both direct
and indirect, represent the single larg-
est technology expenditure each year.
An ERP system is more than just an
information system, howeverit
embodies the institutions business
rules. It must be closely tied to the
institutions business needs for the
institution to realize the systems full
benet.
Therefore, not only is it critical that
institutions choose an ERP system
with utmost care, they should peri-
odically review that system. Such a
review should highlight, to the extent
possible, the total cost of ownership
(TCO) and how well the system aligns
with the institutions business needs.
What follows is a description of just
such a review conducted at Gonzaga
University.
A Brief History of
Gonzagas ERP System
In 1995, Gonzaga University
embarked on a project to implement
a university-wide information system.
The search for an out-of-the-box
solution began following an attempt
to build an integrated data management
system in-house. In the early 1990s,
the university had slowly begun to
digitize its student records as well as its
nancial and student accounts data.
The university realized the advantage of
tying these different data systems into
an integrated system and the efciency
to be gained by managing one set of
computer records. After more than two
years of trying to get different depart-
ments to specify their data needs and
to integrate those needs with those of
other departments, however, database
managers had made little progress.
In 1994, Gonzaga decided to look at
commercial solutions to its database
management problems. With the bless-
ing of the universitys administration,
the CIO pulled together a steering com-
mittee that oversaw the development of
a request for proposal (RFP), the solici-
tation of bids, and the awarding of a
contract. The ERP system selected was a
database-centric suite of software appli-
cations that supported admissions, reg-
istration, nancial aid, nance, human
resources, and university advancement.
The implementation of Gonzagas ERP
solution took two years and was com-
pleted in 1997. Since that time the appli-
cation has matured, with the schools
business processes being melded into
and molded by its use.
Six years have passed since the migra-
EDUCAUSE QUARTERLY Number 3 2005
42
tion to the new ERP system from in-
house developed and third-party soft-
ware solutions. University management
asked the CIO to review the investment
Gonzaga had made in the software and
its relationship with the software ven-
dor. Management also wanted to deter-
mine if the current strategy for use of
enterprise software was still sound or if
a different way of supporting the busi-
ness functions of Gonzaga would be
advantageous and a strategy of change
prudent.
Questions to Answer
This study addressed a number of
questions:
1. What was the total cost of acquiring
and implementing Gonzagas ERP?
2. What is the ongoing TCO?
3. Is it possible to calculate the univer-
sitys return on investment (ROI)?
4. Does Gonzagas ERP meet the major
business needs for which it was
purchased?
5. What are other approaches to sup-
porting the same business functions,
and what would it cost to migrate
to and support these solutions? Are
there other cost-effective ways to
meet the same business needs?
6. What approach do other institutions
take to information management?
What core applications are they
using? What are their experiences
and relationships with their applica-
tion vendors?
Analysis Methodology
Based on the answers to the above
questions, recommendations were to
be made regarding Gonzagas ERP use
and suitability.
Total Cost of Acquisition and
Total Cost of Ownership
Both the total cost to acquire Gonza-
gas ERP system and the total cost for
ongoing support were determined by
combining the direct and indirect costs
for each. Direct costs include hardware,
software, licensing, support contracts,
and consultation. Indirect costs consist
primarily of the salaries of staff neces-
sary to support the system. This includes
employees tasked with supporting the
ERP both within and outside Gonza-
gas central technology department.
In the latter case, the work performed
was often classied as technology-type
work but included making policy and
process decisions necessary for imple-
mentation of the ERP and its operation.
Where necessary, an employees salary
was prorated by the time the employee
spent supporting the system.
Return on Investment
ROI is, conceptually, the savings
returned to the institution by the adapta-
tion of a new business system or process.
Ideally, an institution would be able to
show either that a new business system
was less expensive than the existing sys-
tem or that a new system accomplished
the same work more efciently than the
existing system.
ROI ends up as one of the more elusive
measurements in higher education. First,
while the costs of the new system are easy
to identify, the savings often are not. The
return part of ROI often comes from
indirect savings, which typically do not
appear on the bottom line. Instead they
are the savings that come from operating
more efciently, or even from expenses
avoided. Second, it can be difcult to
clearly attribute some savings to the new
system and notin whole or in partto
other organizational changes. Institutions
of higher education are dynamic, with
myriad changes taking place at any time.
It can be difcult, if not impossible, to
condently assign a savings or a return
to a particular change, especially in a post
hoc analysis.
Business Needs Assessment
The RFP by which the university chose
its ERP served as the vehicle to judge the
extent to which the ERP was meeting
business needs. The heads of various
critical departments were surveyed to
determine the extent to which the ERP
did or did not meet those original busi-
ness needs.
Comparison with Peer
Institutions
The CIOs of nine other universities
were interviewed to determine what
software they used to support their
business functions; their experience
with their software vendor(s); and their
propensity to change applications. This
select group of institutions was reason-
ably comparable to Gonzagaall were
private schools with enrollments of
less than 10,000 students. The results
of this survey were combined with one
conducted by the Association of Jesuit
Colleges and Universities (AJCU). The
AJCU consists of 28 Catholic colleges
and universities of various sizes in the
United States.
Gonzagas CIO kept extensive notes
and les on the process of selecting
and implementing an ERP. These
materials proved invaluable in looking
back to 1995 and 1996. The original
RFP was used as the baseline for proj-
ect goals. The CIOs notes gave a full
picture of the implementation team
and its meeting schedule; the notes
were used to provide an estimate of the
personnel costs for implementation.
Many key department heads also kept
notes of the implementation process
within their departments, and these
notes were used to estimate how the
implementation affected those depart-
ments. Thus, the study was able to
determine, with a comfortable degree
of accuracy, the initial cost of imple-
mentation. Costs for ongoing support
relied on current expenditures and
estimates of personnel time.
Results
To evaluate the Gonzaga ERP imple-
mentation, the study sought data in
each of the categories considered rel-
evant: total costs of acquisition and
ownership, ROI, business needs, and
comparison with peer institutions. The
ndings follow.
It can be difcult,
if not impossible,
to condently assign a
savings or return
to a particular change,
especially in a
post hoc analysis
Number 3 2005 EDUCAUSE QUARTERLY
43
Total Cost of Acquisition
To determine the total cost of acquisi-
tion of the ERP system, the study com-
bined direct and indirect costs.
Direct Costs. Gonzagas original budget
for this project was just under $1 million,
including software, consulting, and the
cost to upgrade to an HP 9000 mid-range
server. The consulting budget was to
cover implementation, education, and
training, as well as some customization
of the ERP.
There were signicant overruns in
direct costs, which added approximately
one-third to the original budget. These
overruns were attributed to
a mid-stream change in the software
from a character-mode screen format
to a graphical user interface (GUI)
screen format;
a lack of understanding of the amount
of Gonzaga resources needed to assist
in the project, which drove a signi-
cant increase in scope and cost from
the application vendor;
a need to engage the application ven-
dor to perform programming tasks
that were assumed Gonzaga staff
would perform; and
a signicant increase in the need for
vendor education of Gonzaga staff.
Indirect Costs. Indirect costs more than
doubled the cost of acquisition. They
were estimated at about $2.2 million,
making the estimated total cost of
acquisition close to $3.5 million. The
majority of those costs were human
resources devoted to implementing
and migrating to the ERP. The cost of
human resources was estimated from
the project plan developed by the CIO.
The CIO closely detailed the makeup
of both managerial and departmental
resource teams, their project meeting
plans and frequency, and the duration
of the project. Discounting current
employee costs developed a historical
cost of resource. Time logs kept by some
department heads during the project
were used to estimate departmental
resource investments in the ERP
project.
Gonzaga anticipated it would take
a signicant investment of time and
other resources to acquire and imple-
ment its ERP. However, the magnitude
of the costs was surprising. If Gonzagas
experience is typical, the indirect cost
to an organization of implementing a
business-wide software application far
exceeds the direct costs. Regardless of
the actual numbers, it is likely that most
institutions fail to understand exactly
how much staff time is required, nor do
they account for the value of that time.
A signicant additional burden is also
placed on the staff, particularly in terms
of data and process migration.
Total Cost of Ownership
Again, TCO was estimated by com-
bining direct and indirect costs of
ownership.
Direct Costs. The direct costs for ongoing
support of Gonzagas ERP were calculated
as the total of annual licensing for the
ERP as well as supporting software, the
costs of supporting hardware, and an
amortization of the mainframe system
that supports the ERP. The hardware
for the ERP is amortized over a ve-
year perioda bit on the long side for
equipment of this type, but it balances
hardware aging versus the demands of a
tight budget. Ongoing costs for software
represent about 20 percent of the initial
software expenditure, which is typical
for enterprise applications. Direct costs
of ownership are about 42 precent of the
total ongoing costs to support the ERP.
Indirect Costs. The total indirect
costs are less well articulated. Included
here are the costs of staff tasked with
supporting and developing the ERP,
whether or not they are in the central
technology department. Also included
are the ongoing costs for internally
driven training. Together, these
indirect support costs are 58 percent of
the total ongoing cost to support the
ERP. As would be expected, however,
the ongoing indirect costs are a mere
fraction of the cost to acquire and
implement the system (18 percent).
See Table 1 for a summary of the total
costs of ownership.
As should be clear, supporting an
ERP demands a signicant investment
of funds. Gonzaga annually expends
close to three quarters of what it cost to
acquire the system to support it. More-
over, a little over half of the costs are
indirect in the form of staff dedicated
to system and end-user support.
Return on Investment
Disappointingly, it proved impossible
to estimate the ROI for the present proj-
ect because no nancial justication
was developed when Gonzaga decided
to move from a homegrown system to
an ERP system. As strange as this seems,
it makes perfect sense. University man-
agement knew that an integrated soft-
ware system was the only way that busi-
ness processes could be standardized
and the foundation built that would
allow Gonzaga to grow and prosper.
Increased efciencies and productivity
were the common theme for ROI from
the ERP system.
ROI discussions with staff quickly
turned to discussions of what it would
cost if Gonzaga did not use an inte-
grated software system. Several people
in the development area stated that
departmental headcount would have
to be increased by six positions. The
admissions department reduced head-
count by four after the ERP implemen-
tation, while enrollment increased by
a factor of two. General consensus was
that because of the ERP capabilities,
Gonzaga was able to maintain level or
reduced stafng while the university
grew substantially.
Table 1
Costs of Ownership
Direct Costs Percent
Software and
licensing
29
Hardware 4
Amortization 9
Indirect Costs
Training 6
Gonzaga staff 52
Total 100
EDUCAUSE QUARTERLY Number 3 2005
44
Business Needs Assessment
We surveyed key department heads
to determine the role the ERP system
plays in the universitys business pro-
cesses. The survey included a look back
at the universitys expectations for the
ERP implementation and asked whether
respondents believed the ERP solution
has met RFP requirements. Department
heads were also asked if they were aware
of other ERP systems or best-of-breed
systems that would better meet their
departments needs or the needs of their
constituents. These needs were the same
used to develop the initial RFP (detailed
in the sidebar).
No one interviewed said that needs
were not being met. When presented
with the list of system expectations
drawn from the RFP, all agreed that the
system met their basic needs. There was
one unanimous exceptionthe need
for an easy to use, ad hoc report writer.
The consensus was that the ERP systems
report writer is difcult to use, espe-
cially if only used occasionally. Once an
employee was experienced in extracting
and reporting data, the tool became
second nature.
A recurring theme hammered home
by everyone interviewed was the value of
an integrated system and a single histori-
cal database for all business and student
information. Several employees stated
that the ERP far exceeded their expec-
tations and that it fully supported the
mission of both the university and their
specic departments. The core reason
for this view is the availability of clean,
reliable data supporting a system that
enforces set business processes. Some of
the supporting comments follow:
Gonzaga could not have progressed
to the point it is today without an inte-
grated application like [our ERP].
We could not be raising the kind
of funds we are today without [our
ERP].
ERP System Expectations Based on RFP
General Objectives
To provide a single integrated univer-
sity information system that
Reduces redundancy and stream-
lines data entry using a variety of meth-
ods such as keyboard entry, scanning,
bar-coding, and others.
Allows current and historical data
to be efciently stored, secured, and
accessed to provide accurate informa-
tion to the university community.
Provides a sophisticated tracking
system for use by all areas of the
university.
Provides an easy-to-use, user
friendly, ad hoc report writer that can
easily access any eld (with appropriate
security) to extract data for producing
management, information, or special
reports on screen, printed, or down-
loaded to the workstation.
Provides a exible system that can
adapt to changes and will continue to
meet our needs in the future through
new functionality, enhancements, and
improvements offered by future soft-
ware releases.
Utilizes a system based on UNIX
and Open Systems standards, which
better poises the university for future
technological decisions.
Integrates with various PC
applications.
Allows for the opportunity to
consolidate computing services for cost,
operating, and technology efciency.
Allows for integration of image
processing, campus-wide ID cards, a
kiosk system, and telephone processing
of information.
Provides the university with the
opportunity to seriously examine the
current processes and determine how
to improve and realign our business
policies and practices in order to better
serve the university community.
Provides opportunities for better
personnel utilization.
Is completely operational, in its
baseline state, by the April/May 1997
timeframe.
Alumni/Development
Provide effective management
of alumni information and events
including maintenance of historical
information.
Provide effective and accurate
tracking of cash gifts, pledges and
outstanding pledges, estates, matching
gifts, and other development records,
allowing for management of histori-
cal data on donors and prospects to
include tracking progress of prospects
through cultivation, solicitation, and
stewardship.
Provide an effective means of man-
aging the gift accounting and acknowl-
edgement process.
Allow for coding of donors/pros-
pects according to region and/or
interest.
Financial
Maintain all required account-
ing records in accordance with GAAP,
NACUBO, the federal government, and
other agencies and standards.
Maintain all accounting records
and les to ensure accurate and ef-
cient budget management, from both a
current and historical perspective.
Ensure that all requirements for
nancial statements, audit reports,
grant reports, IRS compliance audits,
etc., can be met accurately and
efciently.
Number 3 2005 EDUCAUSE QUARTERLY
45
[Our ERP] revolutionized the use of
data in the university environment.
It was critical for Gonzaga to recon-
cile and standardize its business pro-
cesses. [Our ERP] did that for us.
Implementation of [our ERP] was the
single greatest achievement of the last
decade at Gonzaga.
Comparison with Peer
Institutions
What do other colleges and uni-
versities use for their core business
application(s), how do they like their
vendor relationship, and how well are
they being supported? Of the nine insti-
tutions surveyed, 67 percent used the
same ERP system as Gonzaga for their
core application. All of them used enter-
prise applications; no university used a
best-of-breed approach. Of the 28 insti-
tutions in the AJCU, 86 percent of them
use an ERP solution, 11 percent use best-
of-breed, one uses a homegrown appli-
cation, and one uses nothing.
Interestingly, no institution was inter-
ested in changing. Each was pleased
with its vendors support, frequency of
feature/function upgrades, response to
technical issues, and so on. Some com-
plained of being oversold or of being
sold vaporware, but that is a prevalent
phenomenon in the enterprise software
world. Most of the institutions surveyed
have looked at other enterprise appli-
cations from time to time but found
no compelling reason to change. The
benets of changing did not outweigh
the costs.
Of the nine CIOs interviewed, none
would ever again use anything other
than an enterprise-level application.
Implementing a best-of-breed solution
was deemed too costly and, as one CIO
stated, A step back into the dark ages.
Some CIOs were concerned about indi-
vidual small software companies going
out of business, and two had experi-
Ensure that all internal requests for
any type of nancial information (indi-
vidual payroll, departmental revenue or
expenditure, etc.) can be met promptly.
Provide authorized end-user access
to appropriate nancial information
online.
Financial Aid
Provide annual updates of recent
regulatory changes affecting the
administration of the federal student aid
programs.
Provide for enhanced budgetary
controls in order to monitor offers,
acceptances, and declinations of all
types of nancial aid.
Reduce redundancy between the
Student Employment, Financial Aid, and
Payroll databases for student worker
data.
Allow for certain holds to be
created that are document- or fund-
specic in order to allow or prohibit
certain types of nancial aid from being
awarded or disbursed.
Provide for automation of the
receipt and accounting of electronic
fund transfers for student loan proceeds.
Eliminate the majority of manual
efforts involved in monitoring of satis-
factory academic progress.
Allow for incorporation of a nan-
cial aid voice response system.
Provide for automated packaging
of nancial aid based on packaging
philosophies and criteria determined by
our nancial aid management.
Provide for efcient passing of
applicable taxable aid (i.e., waivers)
information to Payroll.
Human Resources
Provide an integrated database
for Human Resources and Payroll to
streamline and enhance management
of employee records.
Provide a source of data on appli-
cants and employees that HR can use
to monitor, manage, and plan various
aspects such as Employment Admin-
istration, Job Classication, System
Design, Total Compensation System
Design, Benets Administration, Equal
Employment and Afrmative Action
Compliance, Position Management,
Position Control, and Employee Rela-
tions Administration.
Provide a source of data for federal
and state compliance reporting.
Provide the ability to maintain
and report historical salary and benets
data by program, department, and/or
employee group.
Student Information
Provide for automated billing and
letter generation functions throughout
the Student system.
Provide for the advancement of
institutional research, through the avail-
ability of centralized data, for retention
and marketing efforts.
Allow restricted access to students
in order to eliminate unnecessary pro-
cessing steps.
Provide for an accurate, stream-
lined process for academic records from
Recruitment through Degree Audit.
Provide services to achieve enroll-
ment goals.
Provide faculty access to advising
information.
Provide for a centralized location
management facility.
EDUCAUSE QUARTERLY Number 3 2005
46
enced just that problem. Most men-
tioned either the difculty or the impos-
sibility of integration as a reason not
to consider best-of-breed systems. Two
also stated that best-of-breed solutions
do not allow for setting and enforcing
business processes, which is critical to
all but the smallest universities.
The study also explored outsourcing
and collaboration as ways for Gonzaga
to obtain the business application sup-
port needed other than by managing
its own ERP.
Outsourcing. With an outsourcing
solution, an outside vendor owns the
equipment and software and provides
connectivity to the application, support
for the application, and other services for
a fee. The systems are usually housed in
a hardened data center. The institutions
connection to both the database and the
application is by a high-speed dedicated
network or through the Internet.
Although we did not price outsourc-
ing as part of our research, this solution
usually costs more. Moreover, institu-
tions usually have some cultural com-
fort issues with outsourcing. First is the
concern that Someone else has my
data! Outsourcing vendors are aware
of this concern and take steps to build
systems and processes to assuage the
nervousness of their prospects. Second
is a concern about the responsiveness
of the vendor to problems or change
requests. There is a general belief that
the closer you are to the resource, the
better the support you will receive,
and the farther away, the worse. How-
ever, responsiveness usually isnt an
issue after the system is turned over
to the outsource provider and things
are stabilized.
Collaboration. With a collaborative
or consortium solution, two or more
universities negotiate collectively to
purchase hardware and software on the
theory that cost per user will be driven
down by economies of scale, particularly
from the standpoint of the server and
storage farm. One university hosts the
equipment and the other consortium
members connect as they would if
the service were outsourced. For those
consortium members not hosting the
service, not only are there the same
set of concerns as in the outsourcing
solution, but additional questions about
data integrity and security and about
the hosting institutions commitment to
equal service. Can non-host institutions
be ensured parity of support, or will the
host institution take care of itself rst?
Many software application vendors are
not supportive of these arrangements.
Consortium solutions often translate
into lower licensing fees and larger
support headaches.
Conclusions
Each university needs a system that
supports all its business functions,
which need to be integrated. Business
processes and practices must be well
dened, supported, and enforced, and
the universitys ERP system is central to
this enforcement.
Universities, in their role as busi-
nesses, need accurate, clean, stable, cur-
rent, and historical data. For universities
to make informed decisions, to operate
efciently, and to offer their students
the best educational experience pos-
sible, they need the best data possible.
An ERP that integrates the business data
on which an institution relies best meets
this need. Moreover, data must be avail-
able in real time to users across multiple
departments and business functions.
University faculty, staff, and students
should nd the system easy to use.
The cost of supporting technology
is high for any business. Moreover,
as the present study demonstrated,
much of the cost of implementing and
maintaining technology can be hidden
from view. Approximately two-thirds of
Gonzagas total cost for implementing
its ERP system was indirect. Still, the
system goes a long way to pay for itself
in terms of savings from unlled staff
positions alone. Even more difcult to
price are the increased efciency and
improved customer service the system
provides.
Gonzaga still has much more to gain
from its ERP implementation. The uni-
versity needs to extend its investment
in the ERP system to support its new
technological vision and fulll the
expectations of its stakeholders. The
university should determine the func-
tionality available within the applica-
tion that is not being used, develop a
program to begin using this unfullled
potential, and educate staff on the
features and use of the added func-
tions. Finally, Gonzaga should develop
an ongoing educational program for
mid to senior management to train
them in new features and refresh their
knowledge of the systems capabilities.
These steps will help Gonzaga Uni-
versity take full advantage of its ERP
implementation. e
Endnotes
1. R. B. Kvavik et al., The Promise and Per-
formance of Enterprise Systems for Higher
Education (Boulder, Colo.: Educause Cen-
ter for Applied Research, Research Study,
Vol. 4, 2002); <http://www.educause
.edu/ers0204/>.
2. K. Pitter and M. Quiner, Building the Raft
While Going Through the Rapids, pre-
sented at EDUCAUSE 2002, Atlanta, Ga.;
PowerPoint slides available at <http://
www.educause.edu/LibraryDetailPage/
666?ID=EDU02119>.
3. G. Spencer, Current Issues Roundtable,
Administrative Systems and Online Stu-
dent Services, roundtable conducted
at EDUCAUSE 2001, Indianapolis, Ind.;
meeting notes available at <http://www
.educause.edu/ir/library/pdf/EDU0195
.pdf>.
Wayne D. Powel (powel@gonzaga.edu) is
Associate Vice President for Information Tech-
nology at Gonzaga University in Spokane,
Washington. Jim Barry is Chief Executive
Ofcer of 360 Consulting Group in Spokane,
Washington.
Business processes
and practices must
be well dened,
supported, and enforced,
and the universitys
ERP system
is central to this
enforcement